This study compares the major Verbal Communication Verbs (VCVs) in English say, tell, speak and talk with their Swedish correspondents säga, berätta, tala and prata. The analysis is based on data from the English–Swedish Parallel Corpus. The semantic and functional description of the verbs is based on the theory of semantic frames and on speech act theory. The verbs are used primarily to report speech, but say and tell and Swedish säga are used also metalinguistically as a commentary on the current discourse as it unfolds. In English, talk and speak turn out to have a wide range of uses that are divided up in a different way in Swedish, whereas tala has many language-specific uses in Swedish. Tell has two major semantic correspondents in Swedish, berätta, which is used to report a complex sequence of events or facts, and the particle verb tala om, which tends to report a single fact. However, tell has a rather general meaning and the most frequent translation is actually säga ‘say’. That tell lacks a direct equivalent in Swedish also explains why tell turns out to be significantly underrepresented in English texts that are translated from Swedish in comparison to original English texts. Genre-based differences are also discussed. For example, not only are say and säga much more frequent in fiction than in non-fiction, but the uses are also distributed differently.
Place, as one of the most basic semantic categories, plays an important role in children’s literature. This contrastive corpus-based study aims to examine and compare how place, in its widest sense, is expressed in children’s literature in English and Czech. The study is data driven and the main methodological approach taken is through n-gram extraction. At the same time, it aims to further test the method, which in previous applications in contrastive analysis has raised a number of methodological issues: while giving reassuring results when applied to typologically closer languages, it proves to be challenging in the study of typologically different languages, such as English and Czech. The second objective of this study is therefore to further address these issues and explore the potential of this methodology. The analysis is based on both comparable and parallel corpora: comparable corpora of English and Czech children’s literature and a parallel corpus of English children’s literature and its translations into Czech.
This paper presents a contrastive analysis of the English noun place and the corresponding Norwegian nouns plass and sted. The study involves two stages. First the patterns of cross-linguistic correspondences of the nouns are established by means of translational data from the English–Norwegian Parallel Corpus. The correspondence patterns reveal both differences and commonalities between the words. Secondly, recurrent lexical bundles involving place, sted, and plass are investigated in order to discover their selectional preferences. The combined approach to the meanings and usage patterns of the words show the two Norwegian nouns to be almost in complementary distribution in many of their patterns. Place is broader in its meaning and shares senses and uses with both of the Norwegian nouns, although in certain contexts sted/plass correspond to different nouns (e.g. room or space), or to adverbs in -where. Certain idiomatic expressions are similar across the languages, but non-literal uses typically do not correspond to a spatial expression in the other language.
At is commonly understood to be one of three basic topological prepositions in English, the other two being in and on. While there are close equivalents in Swedish and Norwegian to both in and on, this is not the case for at. This chapter investigates the choices made by both Swedish and Norwegian translators of physical location predications containing at. It investigates whether the Swedish and Norwegian translation correspondences of the English preposition can aid us in mapping its semantic network. The corpus data for the study comprise all tokens of at coding physical location in the English original fiction texts found in both the English–Swedish Parallel Corpus (ESPC) and the English–Norwegian Parallel Corpus (ENPC). Roughly 25% of these tokens are translated into Swedish and Norwegian by the on preposition (på/på), 25% by the in preposition (i/i) and 25% by the by preposition (vid/ved). Some 12% are translated by other prepositions and the remainder by divergent constructions. The analysis of these translation correspondences leads to the proposal of a semantic network for at.
The present study concerns English hyphenated premodifiers translated into German and Swedish. The material was collected from the fiction part of the English–Swedish Parallel Corpus and the Oslo Multilingual Corpus, and includes almost 700 instances of translations into both German and Swedish, as well as 500 instances each of translations from German and Swedish into English. In the material, hyphenated premodifiers come in many different forms. However, they are mostly short, often containing nominal heads (head-office (man)), ed-participles (water-filled (ditches)) or adjectives (gray-green (tweed)), and only a few are longer, creative hapaxes ((her) “take-me-seriously-or-I’ll-sue-you” (demeanor)). The translations into English contain less variation than English originals, as predicted by translation theory. When the premodifiers are translated into German and Swedish they are often restructured, and only half are translated into German and Swedish premodifiers. German and Swedish premodifying compound adjectives/participles are the most frequent equivalents of English hyphenated premodifiers. More complex English premodifiers are often rendered as postmodifiers in German and Swedish. As could be expected from the preferred noun-phrase structures in German and Swedish, German translations have a (slightly) stronger preference for premodification (e.g., the all-embracing unit → die alles umschließende Einheit), while Swedish (slightly) more often uses postmodifying clauses and prepositional phrases (fifteen-year-old schoolgirls → skolflickor i femtonårsåldern). German and Swedish postmodifiers are very rarely translated into English hyphenated premodifiers.
The paper is concerned with lexical realizations of reportive evidentiality (Boye and Harder 2009; Celle 2009; Wiemer 2007; Aikhenvald 2008; Wiemer 2010b; Boye 2012) across different discourse types and languages. Our aim is to see how language specific the realizations and conceptualization of indirect reportive evidentiality are by contrasting the findings of the analysis of the data collected from various monolingual and parallel corpora. One of the purposes of this contrastive analysis is to find out what kind of correspondence one can expect when dealing with the reportive sub-domain of the linguistic category of evidentiality. The analysis is focused on the hearsay adverbs in English (reportedly, allegedly, supposedly) and Lithuanian adverbials neva ‘allegedly’, tariamai ‘supposedly’, esą ‘allegedly’ as well as their bi-directionally established translation correspondences (comment clauses, complement-taking predicates, as-parentheticals, etc.). The present study is corpus-based and makes use of quantitative and qualitative methods of research. The Lithuanian data have been drawn from the Corpus of Academic Lithuanian (CorALit) and from the spoken, news and fiction sub-corpora of the Corpus of the Contemporary Lithuanian Language (CCLL). The English data have been extracted from the British National Corpus (BYU-BNC). To establish translation correspondences between the items under study, a parallel bidirectional fiction corpus ParaCorpEN-LT-EN, and a collection of translations from English into Lithuanian of EU documents (Glosbe) have been used. Our findings indicate that both sets of adverbials are mainly used in written language (news and academic discourse in English and news discourse and fiction in Lithuanian); however, there is very weak equivalence in their translation correspondences. The question is raised whether the Lithuanian adverbials can be regarded as reportive evidentials.
The study explores overt non-prepositional English translation correspondences of the four most common Czech prepositions, v/ve, na, s/se, and z/ze (‘in, on, with, from’). Some of the divergent counterparts are conditioned lexically or peculiar to one preposition. However, some of the most frequent types appear to be quite systematic and associated with the typological differences between the two languages – inflectional Czech and predominantly analytical English. They reveal the consequences of the word-order principles prevalent in the two languages both at phrasal and clausal level. These are particularly prominent where the Czech adverbial prepositional phrase is paralleled by the English subject noun phrase. The clause-initial position of the subject in English coincides with the unmarked position of the theme, i.e. an element which can convey information on the setting or circumstances of the content of the clause. In English the subject therefore tends to assume ‘adverbial’ semantic roles to a much larger extent than in Czech.
English and Spanish minutes both contain two vocabulary sets, one that codifies the ‘field’ and belongs in a given content area, and another that codifies the discursive practices of the genre ‘minutes’. This paper sets out to explore which multi-word combinations can be identified as genre- and step-specific, and what correspondences can be identified across languages. The study draws on an English–Spanish comparable corpus of meeting minutes, tagged on the rhetorical level. A comparable corpus browser with a basic statistic feature has been used to obtain step subcorpora and WordSmith Tools was used to obtain n-grams within rhetorical steps in each language. N-grams were classified as genre-specific, step-specific, field-related, function-word combination or noise. Empirical findings show that for each rhetorical move, irrespective of text ‘field’, a number of n-grams have become readily associated in each of the languages. Since word choice is determined by genre-bound expectations and by context, selections across languages are not obvious and correspondences show different grams and number of grams.
The paper investigates disciplinary, cultural and genre factors and their influence on citation in research writing. The study is based on literature and linguistics research articles written by Lithuanian and British researchers in their native languages as well as literature and linguistics BA papers written by Lithuanian students in English. The focus of the study is on frequency distribution, syntactic integration and types of citations. The results of the study of research articles confirm clear disciplinary variation in the way citations are employed in research writing. No obvious cultural differences were observed in the analysed expert texts, which points towards discipline as a decisive factor in citational trends in literature and linguistics. Citation trends in linguistic and literature BA papers were similar to a certain extent; however, students who wrote literature papers seem to cite more in the manner of professional writers and thus display more similarity with the conventional citational practices of their field.
The present study investigates the frequency and lexical variation in connector use in argumentative texts in English and Norwegian. The concept of ‘connector’ includes adverbial conjuncts and co-ordinating conjunctions, and the material comprises texts by expert and novice (student) writers. The results indicate that conjunctions are more frequent in English expert texts than in the corresponding Norwegian material, but the opposite tendency is found between the novice writers. In terms of lexical variation, there are two main findings. The first is that much of the observed variation is the result of many items being used only once, and the second is that there are a number of cross-linguistic correspondences in the most frequently used lexical items.
This study compares the major Verbal Communication Verbs (VCVs) in English say, tell, speak and talk with their Swedish correspondents säga, berätta, tala and prata. The analysis is based on data from the English–Swedish Parallel Corpus. The semantic and functional description of the verbs is based on the theory of semantic frames and on speech act theory. The verbs are used primarily to report speech, but say and tell and Swedish säga are used also metalinguistically as a commentary on the current discourse as it unfolds. In English, talk and speak turn out to have a wide range of uses that are divided up in a different way in Swedish, whereas tala has many language-specific uses in Swedish. Tell has two major semantic correspondents in Swedish, berätta, which is used to report a complex sequence of events or facts, and the particle verb tala om, which tends to report a single fact. However, tell has a rather general meaning and the most frequent translation is actually säga ‘say’. That tell lacks a direct equivalent in Swedish also explains why tell turns out to be significantly underrepresented in English texts that are translated from Swedish in comparison to original English texts. Genre-based differences are also discussed. For example, not only are say and säga much more frequent in fiction than in non-fiction, but the uses are also distributed differently.
Place, as one of the most basic semantic categories, plays an important role in children’s literature. This contrastive corpus-based study aims to examine and compare how place, in its widest sense, is expressed in children’s literature in English and Czech. The study is data driven and the main methodological approach taken is through n-gram extraction. At the same time, it aims to further test the method, which in previous applications in contrastive analysis has raised a number of methodological issues: while giving reassuring results when applied to typologically closer languages, it proves to be challenging in the study of typologically different languages, such as English and Czech. The second objective of this study is therefore to further address these issues and explore the potential of this methodology. The analysis is based on both comparable and parallel corpora: comparable corpora of English and Czech children’s literature and a parallel corpus of English children’s literature and its translations into Czech.
This paper presents a contrastive analysis of the English noun place and the corresponding Norwegian nouns plass and sted. The study involves two stages. First the patterns of cross-linguistic correspondences of the nouns are established by means of translational data from the English–Norwegian Parallel Corpus. The correspondence patterns reveal both differences and commonalities between the words. Secondly, recurrent lexical bundles involving place, sted, and plass are investigated in order to discover their selectional preferences. The combined approach to the meanings and usage patterns of the words show the two Norwegian nouns to be almost in complementary distribution in many of their patterns. Place is broader in its meaning and shares senses and uses with both of the Norwegian nouns, although in certain contexts sted/plass correspond to different nouns (e.g. room or space), or to adverbs in -where. Certain idiomatic expressions are similar across the languages, but non-literal uses typically do not correspond to a spatial expression in the other language.
At is commonly understood to be one of three basic topological prepositions in English, the other two being in and on. While there are close equivalents in Swedish and Norwegian to both in and on, this is not the case for at. This chapter investigates the choices made by both Swedish and Norwegian translators of physical location predications containing at. It investigates whether the Swedish and Norwegian translation correspondences of the English preposition can aid us in mapping its semantic network. The corpus data for the study comprise all tokens of at coding physical location in the English original fiction texts found in both the English–Swedish Parallel Corpus (ESPC) and the English–Norwegian Parallel Corpus (ENPC). Roughly 25% of these tokens are translated into Swedish and Norwegian by the on preposition (på/på), 25% by the in preposition (i/i) and 25% by the by preposition (vid/ved). Some 12% are translated by other prepositions and the remainder by divergent constructions. The analysis of these translation correspondences leads to the proposal of a semantic network for at.
The present study concerns English hyphenated premodifiers translated into German and Swedish. The material was collected from the fiction part of the English–Swedish Parallel Corpus and the Oslo Multilingual Corpus, and includes almost 700 instances of translations into both German and Swedish, as well as 500 instances each of translations from German and Swedish into English. In the material, hyphenated premodifiers come in many different forms. However, they are mostly short, often containing nominal heads (head-office (man)), ed-participles (water-filled (ditches)) or adjectives (gray-green (tweed)), and only a few are longer, creative hapaxes ((her) “take-me-seriously-or-I’ll-sue-you” (demeanor)). The translations into English contain less variation than English originals, as predicted by translation theory. When the premodifiers are translated into German and Swedish they are often restructured, and only half are translated into German and Swedish premodifiers. German and Swedish premodifying compound adjectives/participles are the most frequent equivalents of English hyphenated premodifiers. More complex English premodifiers are often rendered as postmodifiers in German and Swedish. As could be expected from the preferred noun-phrase structures in German and Swedish, German translations have a (slightly) stronger preference for premodification (e.g., the all-embracing unit → die alles umschließende Einheit), while Swedish (slightly) more often uses postmodifying clauses and prepositional phrases (fifteen-year-old schoolgirls → skolflickor i femtonårsåldern). German and Swedish postmodifiers are very rarely translated into English hyphenated premodifiers.
The paper is concerned with lexical realizations of reportive evidentiality (Boye and Harder 2009; Celle 2009; Wiemer 2007; Aikhenvald 2008; Wiemer 2010b; Boye 2012) across different discourse types and languages. Our aim is to see how language specific the realizations and conceptualization of indirect reportive evidentiality are by contrasting the findings of the analysis of the data collected from various monolingual and parallel corpora. One of the purposes of this contrastive analysis is to find out what kind of correspondence one can expect when dealing with the reportive sub-domain of the linguistic category of evidentiality. The analysis is focused on the hearsay adverbs in English (reportedly, allegedly, supposedly) and Lithuanian adverbials neva ‘allegedly’, tariamai ‘supposedly’, esą ‘allegedly’ as well as their bi-directionally established translation correspondences (comment clauses, complement-taking predicates, as-parentheticals, etc.). The present study is corpus-based and makes use of quantitative and qualitative methods of research. The Lithuanian data have been drawn from the Corpus of Academic Lithuanian (CorALit) and from the spoken, news and fiction sub-corpora of the Corpus of the Contemporary Lithuanian Language (CCLL). The English data have been extracted from the British National Corpus (BYU-BNC). To establish translation correspondences between the items under study, a parallel bidirectional fiction corpus ParaCorpEN-LT-EN, and a collection of translations from English into Lithuanian of EU documents (Glosbe) have been used. Our findings indicate that both sets of adverbials are mainly used in written language (news and academic discourse in English and news discourse and fiction in Lithuanian); however, there is very weak equivalence in their translation correspondences. The question is raised whether the Lithuanian adverbials can be regarded as reportive evidentials.
The study explores overt non-prepositional English translation correspondences of the four most common Czech prepositions, v/ve, na, s/se, and z/ze (‘in, on, with, from’). Some of the divergent counterparts are conditioned lexically or peculiar to one preposition. However, some of the most frequent types appear to be quite systematic and associated with the typological differences between the two languages – inflectional Czech and predominantly analytical English. They reveal the consequences of the word-order principles prevalent in the two languages both at phrasal and clausal level. These are particularly prominent where the Czech adverbial prepositional phrase is paralleled by the English subject noun phrase. The clause-initial position of the subject in English coincides with the unmarked position of the theme, i.e. an element which can convey information on the setting or circumstances of the content of the clause. In English the subject therefore tends to assume ‘adverbial’ semantic roles to a much larger extent than in Czech.
English and Spanish minutes both contain two vocabulary sets, one that codifies the ‘field’ and belongs in a given content area, and another that codifies the discursive practices of the genre ‘minutes’. This paper sets out to explore which multi-word combinations can be identified as genre- and step-specific, and what correspondences can be identified across languages. The study draws on an English–Spanish comparable corpus of meeting minutes, tagged on the rhetorical level. A comparable corpus browser with a basic statistic feature has been used to obtain step subcorpora and WordSmith Tools was used to obtain n-grams within rhetorical steps in each language. N-grams were classified as genre-specific, step-specific, field-related, function-word combination or noise. Empirical findings show that for each rhetorical move, irrespective of text ‘field’, a number of n-grams have become readily associated in each of the languages. Since word choice is determined by genre-bound expectations and by context, selections across languages are not obvious and correspondences show different grams and number of grams.
The paper investigates disciplinary, cultural and genre factors and their influence on citation in research writing. The study is based on literature and linguistics research articles written by Lithuanian and British researchers in their native languages as well as literature and linguistics BA papers written by Lithuanian students in English. The focus of the study is on frequency distribution, syntactic integration and types of citations. The results of the study of research articles confirm clear disciplinary variation in the way citations are employed in research writing. No obvious cultural differences were observed in the analysed expert texts, which points towards discipline as a decisive factor in citational trends in literature and linguistics. Citation trends in linguistic and literature BA papers were similar to a certain extent; however, students who wrote literature papers seem to cite more in the manner of professional writers and thus display more similarity with the conventional citational practices of their field.
The present study investigates the frequency and lexical variation in connector use in argumentative texts in English and Norwegian. The concept of ‘connector’ includes adverbial conjuncts and co-ordinating conjunctions, and the material comprises texts by expert and novice (student) writers. The results indicate that conjunctions are more frequent in English expert texts than in the corresponding Norwegian material, but the opposite tendency is found between the novice writers. In terms of lexical variation, there are two main findings. The first is that much of the observed variation is the result of many items being used only once, and the second is that there are a number of cross-linguistic correspondences in the most frequently used lexical items.